New Bug Encountered: Minus LIST("") removes Duplicates in a List

Please consider the results in the Expression Assistant:

LIST(1,2,2,3) = 1 , 2 , 2 , 3

LIST(1,2,2,3) - LIST("") = 1 , 2 , 3

- LIST("") is meant to remove empty entries. But not to remove duplicates.

Actually… it should remove duplicates as well. Though you can use UNIQUE(…) as well for duplicates.

So how could one remove blanks in a list, but not remove duplicates?

1 Like

@Fabian - If the list you have is something along the lines of {1, 2, 2, 3, “”, “”}, I am pretty sure I have used -{null} at the end and been able to remove empties no problem. So try that.

But, fair warning, I think I have it set up to grab lookup values and if they don’t exist to put in a null value so it doesn’t generate any errors, but I think it should still work for you.

Thank you Dave,
I din’t know that we can use -{null] in expressions.
Just tried it out. It’s working, but it hat the same side effect as -LIST(""). It removes also duplicates. :frowning_face:

Dang. Well, it was worth the shot.
And yes, I believe anywhere you could use “” you can instead use null, which may help improve readability. Not sure if it could possibly create errors elsewhere though.

1 Like

@Fabian I had another thought. I think I thought of a really annoying way to do it. As long as the list isn’t huge, I think this should work decently.

SPLIT(TRIM(SUBSTITUTE(CONCATENATE([YourList])), ", ", ""))," ")

Try playing around with that.

1 Like

Try Select(YourTable[column], true) - Select(YourTable[column], ISBLANK(column))

Hi Fabian, try this
EXTRACTNUMBERS(CONCATENATE([YourList]))

1 Like

@Bahbus Thank you very much Dave. That’s a solution :blush:

Thanks also to @Heino_Matthee and @Jervz.
But in my real case I’m extracting a list of test fragments (not numbers) from OCRTEXT().

1 Like